Julian Foad wrote:
> Folker Schamel wrote:
>
>> If you worry about fixing a) makes it harder to understand b),
>> then there should be better ways to go.
>
>
> Yes.
>
> In this particular case, an appropriate temporary measure might be to add:
>
> if (that pointer == NULL)
> return svn_error_create ("Invalid status result");
>
> thus avoiding the crash but making it clear that this situation is still
> a bug and needs to be investigated and fixed.
Yes, something like that is what I had in mind.
And it should be added at such locations that it provides
new information compared to the known location reported by Steve.
However, I think if currently no-one is having time to do this,
then this means that currently no-one has time for investigating
bug b) anyway, and then fixing a) in Steve's trivial way
is better than doing nothing.
Which also means that Steve's fix is a good idea anyway,
leaving it open if someone is going to invest time into b) later on.
Or is it a trick to "motivate" Steve into investigating b)
by not fixing a) causing angry tsvn users? ;-)
(Note: Under Windows, explorer crashes are not funny!
The exporter is a part of the OS itself!)
Cheers,
Folker
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Thu Nov 4 16:15:36 2004